The paper introduces the Agentic Compositional Generalization hypothesis, suggesting that reinforcement learning (RL) primarily refines high‑level decision‑making behaviors that orchestrate pre‑trained low‑level skills, rather than teaching new domain‑specific skills from scratch. It proposes River, a training recipe that enhances reward quality by filtering low‑quality synthetic environments and adding process‑level behavior regularization. Using River, RL‑trained agents outperform other open‑source 8B models on four terminal‑agent benchmarks, achieving significant gains with fewer than 30% of the training environments.
By Yihang Yao, Bo Pang, Xuan Phi Nguyen, Ding Zhao, Shafiq Joty, Semih Yavuz
arXiv:2602. 01619v2 Announce Type: replace-cross Abstract: Unsupervised Skill Discovery (USD) aims to autonomously learn a diverse set of skills without relying on extrinsic rewards.
By Seyed Mohammad Hadi Hosseini, Mahdieh Soleymani Baghshah
MLExpert is a 35B Mixture-of-Experts agentic model designed to autonomously perform long‑horizon machine learning engineering (MLE) tasks. It uses a post‑training method that combines hierarchical agentic supervised fine‑tuning with Windowed Interaction RL to learn from intermediate outcomes in complex experimentation workflows. On the MLE‑Bench Lite benchmark, MLExpert raises the average medal rate from 37.9% to 72.7%, matching the performance of larger frontier models and demonstrating generalizable optimization capabilities across related tasks.
By Shangheng Du, Jinxin Shi, Shuaiyu Zhang, Yiming Wu, Junpeng Zhao, Shiyang Feng, Xin Li, Liang He, Xiangchao Yan, Bo Zhang
MLExpert is a 35B Mixture-of-Experts agentic model designed to autonomously conduct long‑horizon machine learning engineering (MLE) tasks. It uses a post‑training approach that combines hierarchical supervised fine‑tuning with Windowed Interaction RL to learn from intermediate outcomes in complex experimentation workflows. On the MLE‑Bench Lite benchmark, MLExpert raises the average medal rate from 37.9% to 72.7%, matching the performance of larger frontier models and demonstrating transferable optimization capabilities to other benchmarks such as NatureBench Lite.
EvolvingBrain introduces an adaptive skill system that enables multi-modal agents to self‑evolve in open worlds. The system comprises Stratified Skill Self‑Evolution, which extracts procedural and problematic patterns from trajectories to build a multi‑granularity skill architecture, and State‑Aware Adaptive Execution, which uses structured memory to guide skill fusion based on the agent’s current state. Experiments on VAB‑Minecraft and DiscoveryWorld show improved success rates and generalization, with code available for public release.
By Runxuan Liu, Jingqi Shi, Baoxin Wang, Honghong Zhao, Haoyuan Xu, Yueqing Sun, Jiafeng Liang, Tao He, Dayong Wu, Qi Gu, Shijin Wang, Ming Liu, Bing Qin
arXiv:2606. 29315v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to take actions in the real world and support human decision-making, yet most agents rely on parametric knowledge, fixed post-training data, retrieval, or search.
By Abhranil Chandra, Sankaran Vaidyanathan, Utsav Dhanuka, Varun Gandhi, Scott Niekum
arXiv:2601. 19810v2 Announce Type: replace-cross Abstract: Unsupervised pre-training can equip reinforcement learning agents with prior knowledge and accelerate learning in downstream tasks.
By Octavio Pappalardo
arXiv:2607. 26784v1 Announce Type: new Abstract: Large language model agents often encounter related yet distinct tasks that share reusable solution patterns.
By Zhiyuan Yao, Yuxin Chen, Zhengxi Lu, Zishan Xu, Yueqing Sun, Yifu Guo, Yuquan Lu, Zhengzhou Cai, Kangning Zhang, Zhuowen Han, Zi-Han Wang, Ziang Ye, Qi Gu, Xunliang Cai, Weiwen Liu, Yongliang Shen
UnifiedPlayers is a cooperative framework that jointly adapts planning, execution, and evaluation for tool-integrated reinforcement learning agents. It consists of a Planning Player that generates tasks, an Execution Player that creates multi-turn trajectories with Python tool calls, and an Evaluation Player that builds executable verifiers, all coordinated by role‑specific rewards under GRPO. The approach outperforms prior baselines on mathematical and general reasoning benchmarks and yields a verifier with high adversarial detection accuracy and more discriminative reward signals.
By Wenjie Liao, Liangjie Zhao, Zehong Cao
AUSO (Action-level Unified Skill Optimization) is a method that unifies skill learning and skill use through a progressive, action-aware optimization process. It starts by jointly learning from teacher guidance and environmental outcomes, then shifts to outcome-based policy optimization, and finally evaluates each action under skill-conditioned and skill-free contexts to strengthen beneficial skill-sensitive actions while suppressing harmful ones. Experiments on ALFWorld, WebShop, and SearchQA demonstrate that AUSO consistently improves agent performance and out-of-distribution generalization compared to competitive baselines.
By Huizu Lin, Chengkai Huang, Tianqi Gao, Tao Huang, Daijiao Liu, Tongxin Li, Xiaoyan Sun, Lina Yao
arXiv:2606. 15306v1 Announce Type: cross Abstract: We envision continually learning agentic systems that become more useful over time: as they encounter sequences of related tasks, they should infer the hidden structure shared across those tasks and use it to improve future decisions.
By Daksh Mittal, Tommaso Castellani, Thomson Yen, Naimeng Ye, Fangyu Wu, Minghui Chen, Tiffany Cai, Emmanouil Koukoumidis, William Zeng, Hongseok Namkoong
arXiv:2606. 29082v1 Announce Type: cross Abstract: Would experience designing faster GPU kernels also help close in on a long-standing open mathematical conjecture?
By Young-Jun Lee, Seungone Kim, Minki Kang, Alistair Cheong Liang Chuen, Zerui Chen, Seungho Han, Taehee Jung, Dongyeop Kang